Cancer is a deadly disease that is complicated by a high degree of cellular heterogeneity, both within a patient and between patients. Current therapeutic agents are not designed to combat such heterogeneity, which ultimately leads to a higher chance of drug resistance and relapse. Smart cancer therapy should therefore be specific, yet able to tackle the intra- and inter-patient diversity displayed by cancerous cells. Engineered T cells are ideal candidates to serve as such smart therapeutics. They can be genetically reprogrammed to detect disease-specific signals, make complex decisions, execute proper responses, and developed into memory T cells. These properties are very difficult, if not impossible, to engineer into traditional therapeutic agents, like small molecules and antibodies. Recent works have shown that patients' T cells can be modified to express a cancer-specific receptor ex vivo and, through adoptive transfer back into the same patient, treat various cancers. Our previous work in engineering signaling control circuits in T cells demonstrates that T cells are amenable to high level genetic reprogramming. Together, these results illustrate the potential of using T cells as smart therapy. Here we propose to develop the next generation of T cell-based cancer therapy to directly address cancer heterogeneity. We will leverage our expertise in synthetic gene circuits and T cell engineering to develop novel gene networks in T cells that control when, where, and which cancer-specific receptors are being expressed, thus reprogramming the spatial and temporal activity of cancer killing T cells. Our first class of synthetic gene networks will address the inter-patient heterogeneity by controlling the activity of engineered T cell in vivo based on how individual patient respond to the adoptive therapy, relying on the administration of small molecules that serves as the trigger of the network. Our second class of gene networks will combat intra-patient heterogeneity by directing T cells to tumors through receptors targeted to a specific cancer antigen, and then non-selectively killing the surrounding tumor cells for a defined duration of time. This allows the elimination of many cancerous cell types simultaneously without causing systemic toxicity. Success from this proposal will result in paradigm-shifting therapies with unprecedented level of flexibility, precision, and personalization. The resulting engineered T cells will be the most sophisticated therapeutic agents ever developed. Furthermore, the genetic circuits and design principle developed here will serve as a general platform that can be combined with any tumor-specific receptors and complement existing adoptive T cell therapy, thus this work will have immediate and broad impact on many cancers. By bringing two emerging fields of science-cell-based immunotherapy and synthetic biology- to cancer research, results from this proposal will fundamentally change cancer treatment.